The solitary pulmonary nodule is a common and challenging problem. Increased use of new high-speed CT scanners, as well as the increasing use of CT as a screening tool for lung cancer has added to the number of new nodules detected annually. In particular, the number of small sub-centimeter nodules has markedly increased. The challenge for radiologists is to differentiate between those nodules that are benign and those that are malignant. Currently no imaging test works well in making this determination for small nodules. A technique that we developed, Early Repeat CT (ERCT) allows us to make estimations of nodule growth rates based on differences in nodule volumes obtained at short temporal intervals. This technique combines high-resolution imaging with advanced image processing methods and the resulting growth rates have been shown to correlate well with malignancy status. During our initial funding period we developed and refined our segmentation and analysis techniques and have collected over 100 cases to analyze. We now have identified several areas which require further development and for which we can make significant improvements. This involves integrating information on pathologic features of tumors as well as information obtained from CT acquisition parameters into the models used in our segmentation approach. In addition, as scanning protocols continue to evolve it will be important to be able to quantitate the degree to which they are comparable in regard to nodule volume determinations and growth assessment. This type of quantitative information can be incorporated into protocols to determine the appropriate time to repeat CT scan. The innovations we propose to address in this grant application will markedly enhance the applicability of ERCT in a more diverse, non-research, clinical setting.